Bill Lam asked:
>
> Does this NVIDIA support IEEE double precision
> floating point?
>
I tried to research this through some web-searching at
the NVIDIA website, and by reading a semi-technical
magazine article about their new 8800 series cards. I
could not determine the answer.
I did find research done at University of North
Carolina Chapel Hill which produced a sorting
algorithm that uses a GPU for both 16-bit and 32-bit
precision IEEE floating point calculation. It was
engineered on an older hardware architecture. Word is
that it has been used to perform stream data mining
and relational database queries.
Further back still, from August 2002, NVIDIA announced
a product ("CineFX" Platform) that features 128-bit
IEEE floating-point color precision throughout the
graphics pipeline. This product allegedly enables
real-time cinematic-quality rendering.
These, and the courting of the scientific market, all
suggest to me that NVIDIA recognizes the value of such
floating-point support. I conjecture that the answer
to your question is 'yes'.
Tracy Harms
____________________________________________________________________________________
Get your own web address.
Have a HUGE year through Yahoo! Small Business.
http://smallbusiness.yahoo.com/domains/?p=BESTDEAL
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
